Abstract: This paper, which is entirely unconditional, proves a sharpened almost-all theorem with fully explicit effective constants for the restricted weighted Goldbach sum R_{a,q}(N) := sum over p1+p2=N, p1 = a (mod q), of (log p1)(log p2), with q >= 1 and gcd(a,q) = 1, whose expected main term is M_{a,q}(N) = C_2 * S(N) * N / phi(q), where C_2 = 0.6601618... is the twin-prime constant and S(N) is the binary singular series.The results are organised around four pillars.(I) A complete character-pair decomposition of the second moment of the error E(N) := R_{a,q}(N) - M_{a,q}(N), extracting the exact diagonal constant G/(2*phi(q)), where G = prod_{p>2}(1 + (p-1)^{-2}) in [1.41320886, 1.41320899] is the Gallagher-Goldston constant.(II) A uniform minor-arc L^4 bound: integral over minor arcs of |S(alpha)|^4 dalpha <= kappa_safe * 2^A * X^3 / (log X)^A, with kappa_safe = 4.40, obtained by combining the complete Vaughan identity with the Bombieri-Vinogradov theorem in integral form, with an explicit derivation of kappa_explicit = C_V^2 * c_{L^2} = 4.004 before applying a rigorous 10% safety margin.(III) The effective almost-all theorem: #{N <= X even : |R_{a,q}(N) - M_{a,q}(N)| > C(A,q) * N * (log N)^{-3}} << X * (log X)^{-A}, with the explicit constant K := 2*C(1,4) <= 3.3624, obtained from C(1,4) <= 1.6812 via a Stechkin-type optimisation.(IV) A Pintz-type exceptional-set bound on {N <= X : R_{a,q}(N) = 0}.Every statement in the main body carries the tag [PROVED]. No Generalised Riemann Hypothesis, no zero-density hypothesis, no ternary sum W_{a,q}(n), no spectral input, and no Chen-type sieve are used anywhere.

Abstract: Heat stress severely constrains wheat productivity, yet the mechanisms underlying thermotolerance remain incompletely understood. In this study, we integrated physiological, biochemical, molecular, and ultrastructural analyses to characterize heat-stress responses in four bread wheat (Triticum aestivum L.) genotypes contrasting in heat tolerance. Membrane thermotolerance was assessed using electrolyte leakage and membrane damage rate under increasing temperature stress, while antioxidant defense was evaluated by measuring the activities of superoxide dismutase, catalase, guaiacol peroxidase, and benzidine peroxidase. Gene expression responses were analyzed by qRT-PCR for DREB, HSP16.9, and compartment-specific SOD isoforms, while HSP16.9 accumulation was validated at the protein level by Western blotting. Heat stress induced progressive membrane destabilization in all genotypes. However, the tolerant genotypes Murov 2 and Zirva 85 maintained greater membrane stability than the sensitive genotypes Aran and Gyzyl bugda. This physiological advantage was accompanied by stronger activation of antioxidant defenses, enhanced induction of DREB and HSP16.9, and a more coordinated expression pattern of FeSOD and MnSOD, indicating integrated redox regulation across chloroplastic and mitochondrial compartments. HSP16.9 protein was undetectable under control conditions but accumulated after heat treatment, confirming its stress-inducible nature and supporting its role in heat-responsive proteostasis. Correlation analysis revealed a coordinated response module linking DREB, HSP16.9, MnSOD, total SOD, BPX, CAT, and GPX. Microscopy further showed that Murov 2 preserved chloroplast, mitochondrial, and mesophyll organization more effectively than Aran under heat stress. Collectively, these findings demonstrate that wheat thermotolerance is governed by the coordinated protection of membranes, redox homeostasis, molecular chaperone systems, and organelle structure, providing potential physiological and molecular targets for breeding heat-resilient wheat cultivars.

Abstract: Toxoplasmosis is a prevalent zoonotic disease worldwide, affecting approximately one-third of the global human population. Primary infection with Toxoplasma gondii during pregnancy can induce miscarriage or congenital infection, leading to irreversible damage to the foetus. Moreover, reactivation of T. gondii infection in immunosuppressed individuals can result in fatal outcomes. No vaccine exists to prevent human disease caused by this parasite. Thus, a vaccine that could induce complete and lasting protection against human toxoplasmosis is an unmet need. In this work, a subunit vaccine, consisting of T. gondii membrane proteins (TGMP) from the T. gondii Me49 strain plus CpG-oligodeoxynucleotide adjuvant (CpG), was tested using BALB/cByJ mice. Intranasal immunisation with TGMP plus CpG (TGMP+CpG) raised TGMP-specific serum IgG and intestinal IgA antibody levels, and parasite-specific IFN-γ-producing CD4+ and CD8+ memory T cells. Dense granule proteins (GRA) 2 and 7, surface antigen (SAG)-related sequences 25, 29B, and 34A, microneme protein (MIC) 10, toxofilin, nascent polypeptide-associated complex (NAC) domain-containing protein and NAC subunit beta were identified as immunogenic proteins in the TGMP. Mice immunised with TGMP+CpG were challenged with T. gondii tachyzoites and showed a significant reduction in the parasitic burden in the peritoneal exudate, spleen, and lungs, compared to mice sham-immunised with CpG alone. Altogether, these results indicate that mucosal immunisation with TGMP plus CpG adjuvant is worth exploring as a vaccination approach to prevent toxoplasmosis.

Abstract: Streptococcus pyogenes and Streptococcus agalactiae are major Gram-positive patho-gens implicated in recurrent and invasive genital infections, and the rise of antimicrobial resistance underscores the need for alternative localized therapies. This study combined molecular docking with a prospective pilot clinical evaluation of an essential-oil-based vaginal capsule formulation intended for localized intravaginal administration. Terpinen-4-ol, isoflavone, and secoisolariciresinol diglucoside (SDG) were analyzed against two bacterial targets - the redox-sensing transcriptional repressor Rex from S. agalactiae and the protein tyrosine phosphatase from S. pyogenes - using the 1-Click Docking platform and the Lamarckian Genetic Algorithm. In parallel, 47 women aged 19-27 years were identified with vaginal and/or cervical colonization or infection caused by S. agalactiae or S. pyogenes, and 34 of them entered a prospective pilot treatment study with once-daily vaginal capsules for 7 days; persistent positive cases received an additional 7-day course. Isoflavone and SDG showed the most favorable interactions against the S. agalactiae target, while SDG also displayed comparatively favorable interaction against the S. pyogenes target. In the clinical pilot cohort, microbiological eradication after completion of therapy reached 91.7% for S. agalactiae and 80.0% for S. pyogenes. The parallel trend between stronger in silico prioritization for the S. agalactiae-directed target and higher clinical eradication in the pilot cohort supports a cautious translational hypothesis, but the absence of a control group, the limited sample size, and the exploratory nature of the clinical dataset require restrained interpretation. Overall, these findings support further controlled studies designed to test whether the computationally prioritized phytocompounds contribute to measurable in vivo benefit within localized antimicrobial strategies.

Abstract: This article investigates how an e‑learning platform and a virtual reality (VR) workshop simulator can be integrated into a traditional craft apprenticeship without displacing workshop‑based learning. Drawing on the Craeft glassblowing Pilot 1 at CERFAV, it reports a two‑phase mixed‑methods study contrasting a Traditional Augmented (TA) group, which used a Craeft e‑learning platform and a VR glassblowing simulator, with a Traditional (T) control group following the standard Certificate of Professional Competence (CPC) programme. Quantitative data from formative assessments and CPC examination results are combined with qualitative feedback, satisfaction surveys, self‑assessment questionnaires, and interviews with apprentices and trainers. In Phase 1, where digital tools were deployed in a separated mode alongside existing instruction, the e‑learning platform was perceived as pedagogically valuable, but effects on assessment outcomes were limited and uneven, with greater score dispersion in the TA group. In Phase 2, redesigned hybrid usage scenarios assigned distinct and complementary roles to the e‑learning platform, VR, and workshop practice within an iterative learning cycle, yielding more consistent advantages for the TA group in cross‑cutting theoretical subjects and reducing variance in their scores. Qualitative analyses show that apprentices adopt a pragmatic stance towards digital tools, using the e‑learning platform primarily for revision and exam preparation and VR for workshop discovery and tool recognition, while maintaining a strong attachment to material practice. The study concludes that, in small, high‑stakes craft VET programmes, the impact of virtual learning environments depends less on their intrinsic properties than on their orchestration within coherent hybrid designs and on trainers’ capacity to align them with authentic tasks and assessment regimes.

Abstract: Climate change is intensifying droughts, heatwaves, and hydrological extremes, increasing crop vulnerability and threatening global food security. This study analyzes the scientific evolution of research on remote sensing-based crop climate vulnerability, with emphasis on temporal, geographical, and thematic patterns. A quantitative, exploratory, descriptive, longitudinal, and retrospective bibliometric approach was applied to 2,343 documents indexed in Scopus between 1985 and 2026. The dataset was processed using Bibliometrix 5.1.1 and VOSviewer 1.6.20 to evaluate productivity, impact, collaboration, and intellectual structure, including Reference Publication Year Spectroscopy (RPYS). Results show a sustained annual growth rate of 4%, with 627 sources, 10,408 authors, an average of 5.17 co-authors per document, 35.3% international collaboration, and 19.93 citations per document. China, the United States, and India lead scientific production, while key journals concentrate dissemination. Thematic analysis highlights the dominance of drought-related studies and the increasing importance of machine learning and cloud-based platforms such as Google Earth Engine. The findings indicate that the field has reached a stage of scientific and technological maturity, transitioning from descriptive monitoring toward predictive and operational geospatial intelligence. However, challenges remain in methodological integration, geographical representation, and the translation of scientific outputs into decision-oriented tools for agricultural adaptation.

Abstract: Purpose: This study examined the acute effects of High-Intensity Interval Training (HIIT) and prolonged endurance training (ET) on heart rate variability (HRV) in elite Greco-Roman wrestlers. A secondary aim was to assess the usefulness of HRV in opti-mizing recovery strategies by monitoring post-exercise changes. Methods: Using a longitudinal crossover design, 13 elite male wrestlers completed two training protocols separated by a 15-day washout period. HRV variables were recorded at baseline, pre-exercise, during training, and 24 hours post-exercise. Data were analyzed with a linear mixed model (LMM) and Bonferroni-adjusted post hoc comparisons. Results: A significant main effect of Timepoint was found for all HRV parameters (SDNN, RMSSD, LF/HF ratio, and overall HRV), indicating marked reductions during exercise fol-lowed by partial recovery after 24 hours. A significant effect of Training Type was ob-served for SDNN. Post hoc analysis showed a significantly greater suppression of overall HRV during HIIT compared to ET (p = .012, Cohen’s d = 0.82). Despite these differences, both protocols demonstrated similar recovery patterns at 24 hours. Conclusion: Both HIIT and ET induced acute decreases in HRV, with HIIT causing a more pronounced decline. Nevertheless, HRV recovery after 24 hours was comparable between the two training modalities.

Abstract: This work deals with the time-domain analysis of asymmetrical faults in three-phase systems. Conventional three-phase analysis provides steady-state solutions for asymmetrical faults. Transient analysis, however, is usually performed by resorting either to oversimplified approximate circuits, or to numerical methods. In this paper, a rigorous analytical methodology based on the time-domain Clarke transformation is presented for the most common asymmetrical faults in three-phase systems. In particular, it is shown that asymmetrical faults result in circuit coupling in the Clarke equivalent circuits. Circuit representation of coupling is also derived in the paper. Coupled equivalent circuits allow rigorous analytical solution of transients in case of asymmetrical faults. The analytical results derived in the paper are validated through proper numerical simulation of faulted radial systems.

Abstract: Methylated flavonoids are abundant phytochemicals in Eucalyptus and are of interest be-cause methylation can alter flavonoid diversity, bioactivity and stability. The enzymes re-sponsible for flavonoid methylation in eucalypts are largely uncharacterised. We used comparative leaf transcriptomics of two species with contrasting flavanone profiles, together with protein-structure-guided candidate selection, to identify prospective O-methyltransferases (OMTs) involved in methylated flavonoid biosynthesis. Five candidate OMTs from E. eugenioides were cloned, heterologously expressed and assayed against a panel of flavonoids and a chalcone precursor. The enzymes showed distinct substrate preferences and regioselectivities. EeOMT1 acted as a broad 7-O-methyltransferase, whereas EeOMT3–EeOMT5 preferentially methylated B and C-ring hydroxyl groups, with differing capacities for sequential methylations at different sites. EeOMT2 was of particu-lar interest because it effectively methylated pinocembrin chalcone to alpinetin chalcone, while only weakly converting pinocembrin to alpinetin. Expression–metabolite analyses across E. eugenioides genotypes supported roles relating to in planta accumulation of 5-O- and 7-O-methylated flavanones, for EeOMT2 and EeOMT1, respectively. These findings support a revised model in which alpinetin biosynthesis proceeds, at least in part, through methylation of a chalcone precursor before flavanone formation. This provides a foundation for elucidating flavonoid methylation pathways and for engineering tailored methylated flavonoids for industrial applications.

Abstract: The aim of the study was to analyse the epidemiological characteristics of confirmed COVID-19 cases in the Tumbes region between March 2020 and December 2022. Method: An ecological design with a descriptive approach was used for the study population, which consisted of 51,421 confirmed cases of COVID-19 registered as confirmed cases of COVID-19 residing in the provinces of Contralmirante Villar, Tumbes, and Zarumilla in the department of Tumbes. Confirmed cases of COVID-19 reported in SISOVID by the Ministry of Health and the registry of confirmed cases of COVID-19 in all age groups were recorded. The sociodemographic data collected was supplemented with information available on the web platform of the National Institute of Statistics and Informatics (INEI). The data were organised into databases, georeferenced and analysed using software such as RStudio, applying epidemiological indicators (incidence, prevalence, mortality and lethality). The spatial and epidemiological data were organised and analysed in a database prepared in Microsoft Excel, in order to consolidate the information through data filtering and organisation. Results: Among the main findings, a decrease in incidence was observed between 2021 (7,840/100,000 inhabitants) and 2022 (5,721/100,000 inhabitants), although prevalence increased from 15% to 20%. The mortality rate fell significantly from 2.57 per 1,000 inhabitants in 2021 to 0.33 in 2022. The highest fatality rates were concentrated among people over 60 years of age with comorbidities, especially men. Conclusion: The pandemic was unevenly distributed, affecting adults between 30 and 59 years of age, populations with pre- existing conditions, and densely populated urban areas the most. Spatial analysis made it possible to identify critical areas for improving the health response. It is recommended to strengthen epidemiological surveillance, improve access to health services in densely populated areas, and prioritise care for vulnerable groups such as older adults and indigenous peoples.

Abstract: The Automobile industry shifts from linear to circular economy for sustainability on a global level with respect to the industrial revolution 5.0, but it faces challenges when establishing circular economy. Circular supply chain implementation is dependent on multiple barriers and enablers, including economic managerial, technological, regulatory and social domains, making it ineffective for single factor solution. The purpose behind this review is to conduct a systematic literature review to develop an understanding how these interconnected barriers and enablers can together shape the circular supply chain implementation and their performance, specifically inside the automotive sector which is still remain a little known. By applying the PRISMA framework on 150 peer reviewed articles, research papers. The research shows that literature focuses on primarily on electric vehicle barriers within developing economies. circular supply chain implementation is governed not only by isolated barriers but by complex systematic interdependencies between enablers as well. This interdependencies are of enablers and barriers can be further classified into economical and financial, managerial and organizational, technological and infrastructure, policy and regularity and market and social. The study shows two systematic patterns, driving the transition technology- policy interdependence and conflicting relationship between large scale production and value extraction. The findings also presented a research agenda focusing on strategic value creation through material streams of automotive electronics, plastics and composites with high potential value and further insights are needed. Circular supply chain as a strategic approach for securing critical material supplies, while policymakers could leverage the use of digital tools as the foundational infrastructure for subsidies allocation and prevent the fraud.

Abstract: Background/Objectives: Acute exercise can influence executive function, but the neurophysiological responses linking exercise to cognitive change remain unclear. Functional near-infrared spectroscopy (fNIRS) provides a feasible method for assessing prefrontal oxygenation during movement-based exercise. This study examined whether prefrontal oxygenated hemoglobin (oxy-Hb) during exercise was associated with subsequent changes in inhibitory control after aerobic and game-based exercise in young adults.Methods: Twenty-four healthy young adults completed aerobic and game-based exercise conditions in a randomized, counterbalanced, within-subject design. The aerobic condition consisted of jogging, whereas the game-based condition consisted of a pickleball-based activity. Exercise intensity was monitored during both conditions. Prefrontal oxy-Hb was recorded during exercise using fNIRS, and inhibitory control was assessed before and after each condition using an Eriksen Flanker task. The primary behavioral outcome was Flanker cost improvement, and the primary fNIRS outcome was mean baseline-corrected prefrontal oxy-Hb during exercise. Results: Exercise intensity was comparable between conditions. Greater mean prefrontal oxy-Hb during exercise was significantly associated with greater improvement in Flanker cost (β = 3.71 ms per 0.01 μM, 95% CI [2.13, 5.30], p < 0.001). Game-based exercise elicited higher mean prefrontal oxy-Hb during exercise than aerobic exercise. No significant condition difference was observed for Flanker cost improvement. Conclusions: Prefrontal oxygenation during exercise was associated with subsequent improvement in inhibitory control. These findings suggest that neurophysiological responses during exercise may account for some between-person variability in acute exercise-related cognitive benefits.

Abstract: Soft robotics is a rapidly evolving field that has attracted significant attention within the scientific community. This review analyzes the main advantages of pneumatic technology in service robots across the different application domains defined by the International Federation of Robotics (IFR). By organizing the literature according to application domains, this work aims to clarify the specific benefits of pneumatic and soft pneumatic solutions in each context. The proposed approach distinguishes between traditional pneumatic solutions and the subsequent emergence of soft robotics, in order to highlight how and to what extent soft technologies have reshaped the design and application scenarios. Particular attention is devoted to the role of materials and recent manufacturing techniques used by researchers to fabricate soft pneumatic robots. Finally, current research trends are discussed, with the goal of identifying key directions for the further development of soft pneumatic service robots.

Abstract: Background: Non-communicable diseases (NCDs) account for the majority of global mortality, yet healthcare systems remain largely oriented toward the treatment of acute conditions. This study examines the structural mismatch between contemporary disease patterns and healthcare system organization. Methods: A narrative analytical review was conducted using secondary data from the Global Burden of Disease (GBD) study and World Health Organization (WHO) reports, supplemented by literature from PubMed, Scopus, and Google Scholar (2000–2026). Findings were interpreted using epidemiological transition theory, health systems analysis, and political economy frameworks. Results: The analysis identifies multiple structural drivers of treatment-oriented healthcare systems, including economic incentives favoring curative services, short-term political decision-making cycles, and the historical dominance of the biomedical model. These factors contribute to systematic underinvestment in prevention, rising healthcare expenditures, and persistent global inequalities in access to medical technologies, as demonstrated during the COVID-19 pandemic. The current model is associated with increasing economic burden and projected losses in global productivity by 2030–2050. Conclusions: The findings indicate that the current healthcare model is structurally misaligned with population health needs. Improving health outcomes and system sustainability requires a reorientation toward prevention, long-term health metrics, and the evidence-based integration of complementary approaches within healthcare systems.

Abstract: Background: Although Diagnostic Reference Levels (DRLs) based on anatomical regions are widely used in Computed Tomography (CT) imaging, a clinical-indication-based approach provides a more accurate representation of daily practice and protocol variation. This study aimed to establish typical radiation doses for common CT clinical indications among adult patients at King Abdulaziz University Hospital (KAUH) in Saudi Arabia. Methods: This retrospective cross-sectional study included 298 adult patients who underwent CT examinations between 2020 and 2025 using two dual-source scanners operating in single- and dual-source modes. Demographic data, acquisition parameters, and radiation dose metrics, including volume CT dose index (CTDIvol) and the dose–length product (DLP) were extracted from scanner consoles. Six clinical indications were analyzed: brain trauma, sinusitis, chest metastases (chest Mets), interstitial lung disease (ILD), abdominopelvic metastases (AbdPel Mets), and hernia. Results: Typical median CTDIvol values in mGy were 36.4 for brain trauma, 3.4 for sinusitis, 4.9 for chest Mets, 5.6 for ILD, 7.2 for AbdPel Mets and hernia. Corresponding DLP values in mGy·cm were 654, 50, 173, 188, 344, and 369, respectively. Brain trauma demonstrated the highest radiation exposure, whereas sinusitis CT showed the lowest. Most values were comparable to or lower than international DRLs. Conclusions: This study provides the first comprehensive clinical-indication-based DRL data in Saudi Arabia beyond anatomical benchmarks, supporting ongoing dose optimization and future national DRL development.

Abstract: Artificial intelligence (AI) systems increasingly depend on multi-stage supply chains that incorporate pre-trained models, third-party datasets, open-source libraries, and automated pipelines, creating an expanding attack surface in which model poisoning, dependency compromise, and provenance manipulation can undermine integrity before deployment. Existing AI governance frameworks—including the NIST AI Risk Management Framework and Secure Software Development Framework—acknowledge supply chain risks but do not define verifiable model provenance or cryptographically durable integrity guarantees. The transition to post-quantum cryptography (PQC) compounds this gap: classical digital signatures used to verify model lineage, dataset integrity, and pipeline attestation will become vulnerable to quantum-enabled forgery within the operational lifetime of many AI systems. This paper synthesizes evidence from policy, standards, and incident sources to characterize the AI supply chain threat landscape and the cryptographic dependencies that the PQC transition disrupts. It proposes three integrated design-science artifacts: a Model Bill of Materials with PQC-safe extensions (MBOM-PQC) defining a verifiable provenance schema; a unified signing and attestation pipeline integrating ML-DSA and hybrid signature modes; and a five-level Supply Chain Assurance Maturity Model (SCAMM) for repeatable organizational evaluation. These contributions provide a structured foundation for AI supply chain integrity in cloud-connected, mission-critical smart systems, ensuring verifiable lineage, authenticity, and trustworthiness through the PQC transition. Empirical validation is deferred to future work.

Abstract: This paper reviews how far national laws in Africa acknowledge the communal lands of communities as their property, as compared to family house and farm plots. This is examined in context of the role which commons ownership by communities could and should play in lessening rural poverty. These include rangelands, forest/bushlands, marshlands and other off-farm resource lands, often with lucrative extractive and non-extractive potentials. These are generally developed without community partnership. Findings suggest resistance to acknowledgement of communal lands as lawfully owned in over half of all 55 states. In effect, governments sustain their colonial designation as unowned wastelands, albeit lawfully used. Yet the one quarter of African states which do now acknowledge community ownership of resource commons also suggest the tide may be turning. There is need to promote this lawful possession is entrenched as a sustained stake in new takings and exploitation of these lands, not merely for fairer compensation for losses incurred, but towards adoption of economic growth path which are more directly inclusive of the rural poor. Inter alia, it is timely for post-2030 sustainable development goals to promote this.

Abstract: As part of conservation efforts for Pinna nobilis, a critically endangered bivalve endemic to the Mediterranean Sea, laboratory programs have been developed to maintain and breed specimens. However, progress in the ex situ conservation of the species remains limited and challenging. This study aims to advance the knowledge required to establish effective reproductive protocols for P. nobilis, specifically focusing on the population in the Mar Menor lagoon, one of the last two surviving populations along the Spanish coast. The first phase of this study involved characterizing the reproductive events in the lagoon. Subsequently, two ex situ reproduction experiences were conducted under conditions designed to replicate the lagoon's natural environment. Three reproductive events were detected in the lagoon between 2019 and 2022, and five successful spawning events occurred across the two ex situ experiences. The conditions for maintenance, maturation, and induction of the individuals are described. In all cases, the percentage of fertilized oocytes released was remarkably high, suggesting internal fertilization, but not self-fertilization, within the pallial cavity. Additionally, ex situ individuals exhibited simultaneous hermaphroditism, with synchronous maturation and alternating release of gametes, effectively preventing self-fertilization. These findings represent a significant step forward in understanding the reproductive biology of P. nobilis and contribute to efforts aimed at ensuring the species’ long-term survival.

Abstract: Global water scarcity due to climate change coupled with environmental pollution from artificial fertilizers threatens sustainable agricultural productivity. To address these challenges, innovative circular and sustainable agricultural practices are being sought after globally. Two promising sustainable approaches are the utilization of biochar coupled treated wastewater (TWW). In our experiment, maize biochar (MB) was locally produced by pyrolysis of maize stover waste in a Top-Lit-Up-Draft (TLUD) biochar kiln and characterized using FTIR, SEM, AAS, UV-vis and adsorption isotherms. Then, the performance of beetroots (Beta vulgaris) in terms of the germination rate, growth rate and yield in the cultivation soils fertilized with MB under drip irrigation with TWW was investigated. Post-hoc analysis (Bonferroni correction, α = 0.01667) revealed that cultivation soils fertilized with MB produced beetroots with significantly longer leaves and larger-heavier bulbs than the control. However, doubling MB fertilization dosage from 10 g to 20 g in the cultivation soils did not yield statistically significant improvement in the average leave height, mass and circumference of beetroot bulbs. Therefore, the findings demonstrated that combining MB with TWW enhanced beetroot growth and yield, highlighting a sustainable and circular agricultural system.

Abstract: Federated learning (FL) enables distributed model training in IoT environments while keeping raw data on local devices. However, protecting model-update exchange is difficult on microcontroller-class devices due to strict latency, memory, and energy constraints. Existing studies often evaluate lightweight cryptography outside complete FL pipelines or on more powerful hardware, leaving its practical overhead on MCU-class devices insufficiently explored. This paper presents an end-to-end, hardware-validated secure framework for exchanging model updates in federated learning on resource-constrained IoT microcontrollers. Implemented on ESP32-based edge devices, the framework combines light-weight block ciphers (SPECK, SIMON, and PRESENT), HMAC-SHA256 for integrity verification, and ECDH-HKDF for session-key establishment. The evaluation assessed latency, throughput, RAM/ROM footprint, and energy consumption. Results show that SPECK provides the lowest overhead (0.13 µs/byte, 8.68 MB/s, 138.3 mJ), SIMON offers intermediate performance (0.41 µs/byte, 1.96 MB/s, 184.9 mJ), and PRESENT incurs the highest computational cost (89.37 µs/byte, 0.011 MB/s, 446.2 mJ). In the CICIoT2023 federated intrusion detection evaluation, the secure model maintained stable convergence and achieved 85.43% accuracy after 20 rounds, remaining close to the centralized baseline. These findings demonstrate the practical feasibility of secure model-update exchange in FL on real IoT microcontrollers and provide hardware-grounded guidance for cipher selection under tight resource budgets.